Installing Additional Packages
Cloudera Data Science Workbench engines are preloaded with a few common packages and libraries for R, Python, and Scala. However, a key feature of Cloudera Data Science Workbench is the ability of different projects to install and use libraries pinned to specific versions, just as you would on your local computer.
Generally, Cloudera recommends you install all required packages locally into your project. This will ensure you have the exact versions you want and that these libraries will not be upgraded when Cloudera upgrades the base engine image. You only need to install libraries and packages once per project. From then on, they are available to any new engine you spawn throughout the lifetime of the project.
You can install additional libraries and packages from the workbench, using either the command prompt or the terminal. Alternatively, you might choose to use a package manager such as Conda to install and maintain packages and their dependencies. For some basic usage guidelines for Conda, see Using Conda with Cloudera Data Science Workbench.
To install a package from the command prompt:
- Navigate to your project's Overview page. Click New Session and start a session.
- At the command prompt in the bottom right, enter the command to install the package. Some examples using Python and R have been provided.
# Install from CRAN install.packages("ggplot2") # Install using devtools install.packages('devtools') library(devtools) install_github("hadley/ggplot2")
# Installing from console using ! shell operator and pip: !pip install beautifulsoup # Installing from terminal pip install beautifulsoup
# Installing from console using ! shell operator and pip3: !pip3 install beautifulsoup4 # Installing from terminal pip3 install beautifulsoup4
(Python Only) Using a Requirements File
For a Python project, you can specify a list of the packages you want in a requirements.txt file that lives in your project. The packages can be installed all at once using pip/pip3.
- Create a new file called requirements.txt file within your project:
- To install the packages in a Python 3 engine, run the following command in the workbench command prompt.
!pip3 install -r requirements.txtFor Python 2 engines, use pip.
!pip install -r requirements.txt
Using Conda with Cloudera Data Science Workbench
Cloudera Data Science Workbench recommends using pip for package management along with a requirements.txt file (as described in the previous section). However, for users that prefer Conda, the default engine in Cloudera Data Science Workbench includes two environments called python2.7, and python3.6. These environments are added to sys.path, depending on the version of Python selected when you launch a new session.
In Python 2 and Python 3 sessions and attached terminals, Cloudera Data Science Workbench automatically sets the CONDA_DEFAULT_ENV and CONDA_PREFIX environment variables to point to Conda environments under /home/cdsw/.conda.
!conda install -y -c conda-forge python=3.6.1 feather-formatTo install a package into the python2.7 environment, run:
!conda install -y -c conda-forge python=2.7.11 feather-format
Note that on sys.path, pip packages have precedence over conda packages.
Creating an Extensible Engine With Conda
- Add the following lines to a Dockerfile to extend the base engine, push the engine image to your Docker registry, and whitelist the new engine for your project. For more details on
this step, see Extensible Engines.
RUN mkdir -p /opt/conda/envs/python2.7 RUN conda install -y nbconvert python=2.7.11 -n python2.7Python 3
RUN mkdir -p /opt/conda/envs/python3.6 RUN conda install -y nbconvert python=3.6.1 -n python3.6
- Set the PYTHONPATH environmental variable as shown below. You can set this either globally in the site administrator dashboard, or for a specific project
by going to the project's Python 2